Even Simultaneous Bilinguals Do Not Reach Monolingual Levels of Proficiency in Syntax

Abstract

While there is no doubt that children raised bilingual can become extremely proficient in both languages, theorists are divided on whether bilingualism is effectively monolingualism twice (the “Two Monolinguals in One Brain” hypothesis) or differs in some fundamental way from monolingualism. A strong version of the “Two Monolinguals” hypothesis predicts that bilinguals can achieve monolingual-level proficiency in either (or both) of their languages. Recently, Bylund and Abrahamsson argued that evidence of lower syntactic proficiency in simultaneous bilinguals was due to confounds of language dominance; when simultaneous bilinguals are tested in their primary language, any difference disappears. We find no evidence for this hypothesis. Meta-analysis and Monte Carlo simulation show that variation in published results is fully consistent with sampling error, with no evidence that method mattered. Meta-analytic estimates strongly indicate lower syntactic performance by simultaneous bilinguals relative to monolinguals. Re-analysis of a large dataset (N = 115,020) confirms this finding, even controlling for language dominance. Interestingly, the effect is relatively small, challenging current theories.